Abstract: The present invention discloses a combined oxygen, nitrogen and hydrogen joint determination device and method based on thermal conductivity-infrared technique. In the device, a melting and extraction system comprises a carrier gas introduction pipeline, a chamber valve, an impulse furnace and a dust filter; an analytical gas transmission system comprises a flush valve, a bypass valve, a mass flowmeter and a first switching valve; a signal detection system comprises an infrared cell, an analysis purification reagent tube and a thermal conductivity cell; and a signal analysis system comprises a controller and a processor. The present invention realizes different joint determination modes by switching different gas paths, which can meet different determination requirements of different users.

Abstract: Systems and methods of the present invention provide for: receiving a digital image data; modifying the digital image data to reduce a width of a feature within the digital image data; executing a dimension reduction process on the feature; storing a feature vector comprising: at least one feature for each of the received digital image data, and a correct or incorrect label associated with each feature vector; selecting the feature vector from a data store; training a classification software engine to classify each feature vector according to the label; classifying the image data as correct or incorrect according to a classification software engine; and generating an output labeling a second digital image data as correct or incorrect.

Abstract: Systems and methods are provided for ride order dispatching. Such method may comprise obtaining information on a location of a vehicle and a time to input into a trained neural network algorithm; and based on a policy generated from the trained neural network algorithm, obtaining action information for the vehicle, the action information comprising: staying at a current position of the vehicle, re-positioning the vehicle, or accepting a ride order.

Abstract: The present invention discloses a low-yield-ratio ultra-high-strength high-toughness steel for pressure hulls and a preparation method therefor, wherein the chemical components by weight percentage are: 0.05%-0.10% of C, 0.15%-0.35% of Si, 0.60%-1.00% of Mn, 0.10%-0.50% of Cu, 0.10%-1.00% of Mo, 0.40%-0.70% of Cr, 0.05%-0.15% of V, 5.00%-10.00% of Ni, and the balance of Fe and unavoidable impurities. The technical solution of the present invention adopts secondary quenching heat treatment, the first quenching is performed to achieve complete austenitizing, and then the second quenching and tempering are performed to finally obtain complex phase structures such as tempered martensite, critical ferrite and retained austenite, so as to meet the performance index requirements of low yield ratio, ultra-high strength and high toughness, and thereby promoting application in practice.

Abstract: An ultrahigh-strength ultrahigh-toughness and low-density dual-phase lamellar steel plate is disclosed. The steel plate comprises the following alloy components in percentage by mass: 0.200-0.320% of C, 0.600-2.000% of Mn, 0.200-0.600% of Si, 2.000-4.000% of Al, 0.300-1.200% of Ni, 0.001-0.005% of B, P not greater than 0.012%, S not greater than 0.005%, and the balance of Fe and inevitable impurities. The steel plate consists of dual phases of ferrites and martensites, the ferrites are high-temperature delta ferrites, the martensites are lath martensites, the delta ferrites are distributed in the lath martensites in a lamellar mode. The steel plate has excellent mechanical properties, for example, the yield strength in the rolling direction is not less than 1000 MPa, the tensile strength is not less than 1600 MPa, the elongation is not less than 8.0%, and the average value of Charpy V-Notch impact energy at ?40° C. is not less than 350J.

Abstract: Systems, methods, and computer readable media directed to interactive reinforcement learning with dynamic reuse of prior knowledge are described in various embodiments. The interactive reinforcement learning is adapted for providing computer implemented systems for dynamic action selection based on confidence levels associated with demonstrator data or portions thereof.

Abstract: An on-line solution heat treatment process for austenitic stainless steel plates is disclosed, which comprises continuous casting billet heating, high pressure water dephosphorization, rough rolling, finish rolling, softening process, ultra fast cooling, straightening, cutting to length, shot peening, pickling. The process of the present invention makes full use of the residual rolling temperature, without the need of reheating steel plates from the room temperature to the solid solution temperature. Compared with the conventional steel plate rolling process, the process needs to perform the softening process on the steel plates after rolling to ensure that the deformed austenite is fully softened before cooling. After the softening process, ultra fast cooling is performed to cool the steel plates from above 800° C. to below 430° C. so as to prevent intergranular chromium depletion caused by a large amount of precipitation of high chromium carbides from affecting the stain resistance of stainless steel.

Abstract: The present invention discloses a 785 MPa level extra-thick quenched and tempered rack steel plate for offshore platforms and a preparation method therefor, wherein chemical components of the steel plate by weight percentage are: 0.08%-0.15% of C, 0.60%-1.5% of Mn, 0.10%-0.50% of Si, 0.002%-0.005% of S, 0.005%-0.015% of P, 2.7%-4.0% of Ni, 0.40%-1.00% of Cr, 0.30%-0.70% of Mo, 0.10%-0.40% of Cu, 0%-0.080% of Nb, 0%-0.10% of V, 0%-0.08% of Ti, 0.02%-0.08% of Al, 0.0008%-0.0015% of B, and the balance of Fe and other unavoidable impurities. The present invention adopts a low C, high Ni—Cr—Mo alloying and Nb—V—Ti microalloying component system, has a yield strength of ?795 MPa, a tensile strength of ?865 MPa, a ?60° C. Charpy V-notch impact energy of ?140 J, a Z value of ?35%, and a maximum thickness of 100 mm, and can be used to manufacture racks for offshore platforms.

Abstract: The present invention discloses a low-yield-ratio ultra-high-strength high-toughness steel for pressure hulls and a preparation method therefor, wherein the chemical components by weight percentage are: 0.05%-0.10% of C, 0.15%-0.35% of Si, 0.60%-1.00% of Mn, 0.10%-0.50% of Cu, 0.10%-1.00% of Mo, 0.40%-0.70% of Cr, 0.05%-0.15% of V, 5.00%-10.00% of Ni, and the balance of Fe and unavoidable impurities. The technical solution of the present invention adopts secondary quenching heat treatment, the first quenching is performed to achieve complete austenitizing, and then the second quenching and tempering are performed to finally obtain complex phase structures such as tempered martensite, critical ferrite and retained austenite, so as to meet the performance index requirements of low yield ratio, ultra-high strength and high toughness, and thereby promoting application in practice.

Abstract: Systems and methods for adaptive assessment may utilize test deliver policies derived from a multi-armed bandit approach in combination with an item response theory model. A monotonic policy may involve defining a difficulty range according to which test items are selected for delivery during an assessment, where the difficulty range is updated following delivery of each test item based on test-taker performance. A multi-stage policy may implement several stages for test item delivery, with a different initial difficulty range for test item selection being defined for each stage and the difficulty ranges being updated based on test-taker performance within each respective stage. A probability matching policy may involve defining an item difficulty probability distribution according to which test items are selected for delivery to a test taker, where the probability distribution is initialized based on test taker skill level and updated based on test taker performance.

Abstract: Systems and methods are provided for ride order dispatching. Such method may comprise obtaining information on a location of a vehicle and a time to input into a trained neural network algorithm; and based on a policy generated from the trained neural network algorithm, obtaining action information for the vehicle, the action information comprising: staying at a current position of the vehicle, re-positioning the vehicle, or accepting a ride order.

Abstract: An ultrahigh-strength ultrahigh-toughness and low-density dual-phase lamellar steel plate is disclosed. The steel plate comprises the following alloy components in percentage by mass: 0.200-0.320% of C, 0.600-2.000% of Mn, 0.200-0.600% of Si, 2.000-4.000% of Al, 0.300-1.200% of Ni, 0.001-0.005% of B, P not greater than 0.012%, S not greater than 0.005%, and the balance of Fe and inevitable impurities. The steel plate consists of dual phases of ferrites and martensites, the ferrites are high-temperature delta ferrites, the martensites are lath martensites, the delta ferrites are distributed in the lath martensites in a lamellar mode. The steel plate has excellent mechanical properties, for example, the yield strength in the rolling direction is not less than 1000 MPa, the tensile strength is not less than 1600 MPa, the elongation is not less than 8.0%, and the average value of Charpy V-Notch impact energy at ?40° C. is not less than 350J.

Abstract: An improved machine learning system is provided. For example, a content management server may provide real-time analysis of a user's handwriting to assess the user's knowledge of a language, including using a convolution neural network method. The convolution neural network method may be executed to normalize at least some identified strokes in the user's handwritten user input. Normalization may be performed by translating a window comprising a subset of pixels in a digital representation of the handwritten user input amongst a plurality of pixels in the digital representation.

Abstract: Systems and methods of the present invention provide for: receiving a digital image data; modifying the digital image data to reduce a width of a feature within the digital image data; executing a dimension reduction process on the feature; storing a feature vector comprising: at least one feature for each of the received digital image data, and a correct or incorrect label associated with each feature vector; selecting the feature vector from a data store; training a classification software engine to classify each feature vector according to the label; classifying the image data as correct or incorrect according to a classification software engine; and generating an output labeling a second digital image data as correct or incorrect.

Abstract: Systems and methods of the present invention provide for: receiving a digital image data; modifying the digital image data to reduce a width of a feature within the digital image data; executing a dimension reduction process on the feature; storing a feature vector comprising: at least one feature for each of the received digital image data, and a correct or incorrect label associated with each feature vector; selecting the feature vector from a data store; training a classification software engine to classify each feature vector according to the label; classifying the image data as correct or incorrect according to a classification software engine; and generating an output labeling a second digital image data as correct or incorrect.

Abstract: An improved machine learning system is provided. For example, a content management server may provide a digital assessment of a user's handwriting to assess the user's knowledge of a language. The assessment may comprise adaptive technology to help determine initial questions to provide to the user as well as follow-up questions to clarify appropriate remediation content in a particular context. The content management server may also provide real-time analysis, including assessing multiple users at the same time in adjusting the assessment based on the digital input from each of these users. In some examples, the content management server may incorporate handwriting analysis methods to perform object detection and score handwriting input.

Abstract: A control method for a roller quenching process of a heavy-piece weight and large-section ultra-heavy plate has a specific heat model, heat transfer coefficient model, temperature field model and correction model. Plate parameters inputted include thickness, length and carbon content, technological procedure, roller speed and acceleration. Measured parameters include tapping temperature, temperature after air cooling and temperature after self-tempering. The temperature field model is used. Specific heat model and the heat transfer coefficient model are invoked for calculating an air cooling stage, water cooling stage and self-tempering stage in sequence. Temperature fields are corrected through the correction model. Simulated results include a group of cooling curves and cooling speed curves at different thicknesses. Practical temperature drop curves and cooling speed curves are obtained in combination with actual production and part of actual debugging process is replaced by model calculation.

Abstract: Systems and methods of the present invention provide for: receiving a digital image data; modifying the digital image data to reduce a width of a feature within the digital image data; executing a dimension reduction process on the feature; storing a feature vector comprising: at least one feature for each of the received digital image data, and a correct or incorrect label associated with each feature vector; selecting the feature vector from a data store; training a classification software engine to classify each feature vector according to the label; classifying the image data as correct or incorrect according to a classification software engine; and generating an output labeling a second digital image data as correct or incorrect.

Abstract: Systems, methods, and computer readable media directed to interactive reinforcement learning with dynamic reuse of prior knowledge are described in various embodiments. The interactive reinforcement learning is adapted for providing computer implemented systems for dynamic action selection based on confidence levels associated with demonstrator data or portions thereof.

Abstract: An improved machine learning system is provided. For example, a content management server may provide real-time analysis of a user's handwriting to assess the user's knowledge of a language, including using a convolution neural network method.